Virtual Marathon Training Pacer
Introduction to virtual marathon pacing
A marathon is decided by arithmetic long before it is decided by willpower. The distance is fixed at exactly 42.195 kilometres by World Athletics, so a goal finish time is nothing more than an average pace multiplied by that distance. What separates a controlled marathon from a painful one is whether the runner knows, kilometre by kilometre, what that average pace actually feels like and what the clock should read at each marker. This pacer converts a goal time into an exact split sheet, into the equivalent race times you should be capable of at shorter distances, and into the training paces those numbers imply.
The word virtual in the title matters. A certified road marathon is measured by an accredited measurer riding a calibrated bicycle fitted with a Jones counter, along the shortest route a runner could legally take, with a deliberate 0.1 % short course prevention factor built in so the course can never come up short on re-measurement. A virtual marathon run from your front door and recorded by a wrist GPS has none of that. Satellite geometry, tree cover, tall buildings and the smoothing algorithm in the watch all push the recorded distance around, typically by a few tenths of a percent and sometimes by more than one percent. That is why this tool exposes a distance allowance input: pacing a virtual marathon well means planning for a route that is slightly longer than the nominal distance rather than discovering the shortfall at the finish.
Everything below is deterministic arithmetic on numbers you supply, plus one empirical relationship, the Riegel endurance equation, which is clearly flagged wherever it is used. No physiological testing is simulated, no injury risk is predicted, and nothing here is medical advice.
How to use the pacer, field by field
Start with plan basis. Choose goal marathon time if you already know the finish time you are chasing; choose recent race result if you would rather have the goal derived for you. In the second case you enter a distance you have genuinely raced and the time you ran, and the marathon target is projected from it with the Riegel relation. A recent half marathon is by far the best input; a parkrun 5K is the weakest, for reasons set out under limitations.
Enter times as hours, minutes and seconds in three separate fields so there is no ambiguity about whether 3.45 means three hours forty-five or three and forty-five hundredths of an hour. Minutes and seconds are each capped at 59; if you type 75 minutes the calculator tells you so instead of silently accepting it. Zero-length and negative times are rejected outright, and a goal beyond 24 hours is refused because the Riegel relation has no meaning that far outside its fitted range.
Pace units switches every output between minutes per kilometre and minutes per mile, and switches the split sheet between 42 kilometre segments plus a 195 metre run-in and 26 mile segments plus a 385 yard run-in. The conversion uses the exact international mile, 1.609344 kilometres, not a rounded 1.61.
Split strategy chooses how the effort is distributed. Even splits hold one pace throughout. The negative-split options ramp the pace linearly so that the second half is run one, two or three percent faster in elapsed time than the first; the fade option does the reverse. In every case the total finish time is held exactly constant, so you can flip between strategies and compare the split sheets line for line.
Distance allowance adds a percentage to the planned distance. Leave it at 0 % for a certified course where the blue line is marked for you. On a virtual course, 0.5 % to 1 % is a realistic hedge against running wide on corners and against GPS drift; the calculator then paces you for the longer distance so that you cross the nominal 42.195 kilometres comfortably inside the goal.
Press Calculate pace plan. You can then copy the plan as text, download it as CSV for a spreadsheet or watch-workout builder, or copy a permalink that reloads the page with the same inputs.
The pace, split and Riegel equivalence formulas
The base quantity is average pace, the elapsed time divided by the planned distance:
Formula: p ¯ = T / (D_0 (1 + a / 100))
Here T is the goal time in seconds, D0 is 42.195 km and a is the distance allowance in percent. Speed is simply the reciprocal, and the conversion between units is exact:
Formula: p_mile = 1.609344 p_km, v = D / T
For a non-even strategy the instantaneous pace is ramped linearly in the fraction of the distance completed, x, about the average:
Formula: p(x) = p ¯ [1 + k (1 / 2 − x)]
Integrating that from the start to any point x gives the cumulative clock, which is what a split sheet actually needs:
Formula: C(x) = T [x (1 + k / 2) − (k x^2) / 2]
Because C(1) equals T for every value of k, the finish time is preserved exactly. The ramp coefficient k is chosen so that the second half takes s percent less time than the first:
Formula: k = s / (50 − s / 4), T_2 / T_1 = 1 − s / 100
Race equivalence uses the endurance relation Peter Riegel published in American Scientist in 1981. Fitting a power law to world-record performances across distances, he obtained a fatigue exponent of 1.06:
Formula: T_2 = T_1 (D_2/D_1)^1.06
An exponent greater than one is exactly the statement that pace decays with distance. Doubling the distance costs 21.06, about 2.085 times the time, not twice; a half marathon equivalent therefore lands at 0.51.06 = 47.96 % of the marathon time rather than 50 %, which is why the Riegel half in the table below is always faster than an even half split.
Threshold pace is derived from the same relation by asking what distance the reference performance implies for an all-out effort lasting exactly one hour, then dividing that distance into 3600 seconds:
Formula: D_T = D_ref (3600/T_ref)^1/1.06
Reference table: pace bands across the marathon range
The table compares common goal times. Every column is computed from 42.195 km and the exact mile, and the last two columns contrast an even half split with the Riegel-equivalent half marathon a runner of that marathon standard should be able to race.
| Goal finish | min/km | min/mile | km/h | mph | Even half split | Riegel half equivalent |
|---|---|---|---|---|---|---|
| 3:00:00 | 4:16 | 6:52 | 14.06 | 8.74 | 1:30:00 | 1:26:20 |
| 3:15:00 | 4:37 | 7:26 | 12.98 | 8.07 | 1:37:30 | 1:33:32 |
| 3:30:00 | 4:59 | 8:01 | 12.06 | 7.49 | 1:45:00 | 1:40:43 |
| 3:45:00 | 5:20 | 8:35 | 11.25 | 6.99 | 1:52:30 | 1:47:55 |
| 4:00:00 | 5:41 | 9:09 | 10.55 | 6.55 | 2:00:00 | 1:55:07 |
| 4:30:00 | 6:24 | 10:18 | 9.38 | 5.83 | 2:15:00 | 2:09:30 |
| 5:00:00 | 7:07 | 11:27 | 8.44 | 5.24 | 2:30:00 | 2:23:53 |
| 5:30:00 | 7:49 | 12:35 | 7.67 | 4.77 | 2:45:00 | 2:38:17 |
Worked example: pacing a 3:45:00 goal marathon
Take a runner targeting 3 hours 45 minutes on a certified course, so T = 13 500 s and the allowance is 0 %. Average pace is 13 500 / 42.195 = 319.94 s/km, which the pacer prints as 5:20 per kilometre. In imperial units the marathon is 42.195 / 1.609344 = 26.218757 miles, so the pace is 13 500 / 26.218757 = 514.90 s/mile, or 8:35 per mile. Average speed is 11.25 km/h, equivalently 6.99 mph.
With even splits the 5 km marker should read 26:40, the 21 km marker 1:51:59, the halfway point at 21.0975 km reads 1:52:30, and the final 195 m segment takes 62.4 s. In mile mode the 26 mile marker reads 3:43:07 and the closing 385 yards take 112.6 s.
Switch to a two percent negative split and the ramp coefficient becomes k = 2 / (50 − 0.5) = 0.040404. The opening kilometre slows to 5:26, the twenty-first kilometre passes through at 5:20, and the forty-second is run at 5:14. The first half now takes 1:53:38 and the second 1:51:22 — a ratio of exactly 0.98, and a total that is still 3:45:00 to the second.
Riegel equivalents from that 3:45:00 marathon come out at 23:28 for 5 km, 48:55 for 10 km, 1:21:00 for 10 miles and 1:47:55 for the half marathon. Solving the threshold expression gives a one-hour racing distance of 12.13 km and a threshold pace of 4:57 per kilometre, or 7:58 per mile. Easy running for this athlete sits between 6:08 and 6:56 per kilometre.
Finally, add a 1 % distance allowance for a virtual course. The planned distance becomes 42.617 km and the required average pace tightens to 5:17 per kilometre. That is the honest cost of a route you have not had measured: about three seconds per kilometre.
Reading the split sheet and the training zones
The split sheet is a checking instrument, not a metronome. Its value is that it tells you within a second or two what your watch should read at each marker, so a 15-second drift is visible at 10 km instead of at 32 km when there is nothing left to do about it. If two consecutive segments run more than about ten seconds per kilometre fast, you are burning the reserve that the last 10 km will demand.
The equivalence table answers a different question: is the goal internally consistent with your current fitness? If your best recent 10 km is two minutes slower than the equivalence table's 10 km entry, the marathon goal is ahead of your demonstrated speed, and the gap has to be closed in training rather than on race day.
The zone table names the five training intensities in Jack Daniels' vocabulary. Easy running builds the aerobic base and should feel conversational. Marathon pace rehearses race rhythm and fuelling. Threshold work is comfortably hard and clears lactate at the rate it is produced. Interval work stresses maximal oxygen uptake and is run in repetitions of three to five minutes. Repetition work is short, fast and fully recovered, and it sharpens economy rather than endurance. The great majority of weekly volume for a marathon build belongs in the easy and marathon bands.
Limitations, assumptions and where the numbers stop being trustworthy
These are the assumptions the arithmetic makes, stated plainly.
Riegel is an empirical fit, not a law. The exponent 1.06 was regressed on world-record performances, where every distance is contested by a specialist. It is generally quoted as reliable for efforts between roughly 3.5 minutes and about four hours, which puts a 4:30 or 5:00 marathon outside the range it was fitted over. Extrapolating up from a short race is the worst case: a 5 km to marathon projection is an eightfold jump and assumes long-run volume, glycogen storage and fuelling practice the equation knows nothing about. In practice the equation over-predicts performance for that extrapolation, meaning it hands you a marathon time faster than most recreational runners actually run. Treat a 5K-based marathon target as a ceiling; prefer a half marathon within the last six weeks as the reference performance.
A single exponent cannot describe every runner. High-mileage runners hold pace better over long distances than the 1.06 exponent suggests, and low-mileage runners hold it worse. The same input therefore carries a wider error band for a beginner than for an experienced marathoner, and the calculator has no way to know which you are.
The split model assumes a flat, still, temperate course. Hills, wind, heat, humidity, altitude and underfoot conditions are all excluded. A steady climb, a headwind or a warm day will all cost you time at the same effort, and the correct response is to run the effort rather than the number on the sheet.
Distance is only as good as the measurement. On a certified course the 42.195 km is guaranteed to within 0.1 %. On a virtual course it is not, and the allowance input is a hedge, not a correction: it cannot tell you how far your watch is out on a given day.
Training zones are conventions, not measurements. Marathon pace is definitional and threshold pace follows from the Riegel relation, but the easy band used here (1.15 to 1.30 times marathon pace) is a coaching convention drawn from the easy-running zone in Daniels' work, not a physiological constant measured on you. A laboratory lactate or gas-exchange test is the only way to establish your true thresholds.
Gradual progression is prudent but not proven protective. The familiar rule of increasing weekly volume by no more than ten percent has been tested directly: a randomised controlled trial of 532 novice runners found no difference in running-related injury incidence between a graded thirteen-week programme built on the ten percent rule and a standard eight-week programme. Sensible progression is still good practice; it is simply not the injury shield it is often sold as.
This is not medical advice. The calculator does not know your health history, and nothing here should be used in place of guidance from a qualified clinician or coach. If you are new to vigorous exercise, general activity guidance from the American College of Sports Medicine and the US physical activity guidelines is the right starting point.
Questions runners ask about virtual marathon pacing
Is a virtual marathon the same distance as a certified marathon?
The nominal distance is the same 42.195 km, but the measurement control is not. World Athletics requires a certified road course to be measured with a calibrated bicycle along the shortest possible route an athlete could run, with a short course prevention factor of 1.001 built in and a permitted measurement uncertainty of only 0.1 percent, which is 42 m over a marathon. A virtual marathon recorded by a GPS watch has none of that control, and consumer GPS drifts under tree cover and between tall buildings. Adding a small distance allowance is the only way to be confident you actually covered the full distance.
Why does the Riegel prediction from my 5K look too fast for a marathon?
Riegel fitted the exponent 1.06 to world-record performances, where every distance on the curve is run by an athlete fully specialised for it. Extrapolating from 5 km to 42.195 km is an eightfold jump in distance, and it silently assumes you already have the long-run volume, glycogen stores and race-day fuelling practice needed to hold the scaled pace for three hours or more. Most recreational runners finish slower than a 5K-based marathon prediction, so treat it as a ceiling rather than a target and prefer a recent half marathon as the input.
Should I run even splits or negative splits over 42.195 km?
An even effort is the sensible default, because pace variation early in a marathon costs more time in the closing kilometres than it saves at the moment you bank it. A negative split of one to two percent, meaning the second half is covered one to two percent faster than the first, leaves a margin for the adrenaline surge at the start and is the pattern behind most well-executed marathons. This pacer holds the total finish time fixed whichever strategy you choose, so the split tables can be compared line by line.
What does the threshold pace in the training zone table represent?
Threshold pace is the pace you could sustain in an all-out race lasting about sixty minutes. This calculator derives it by solving the Riegel relation for the distance you would cover in exactly 3600 seconds and then dividing that distance into the hour. It is a comfortably hard effort rather than a race effort, and it is normally trained as a sustained tempo run of twenty to forty minutes, or as cruise intervals of five to ten minutes with short jogged recoveries.
How precise is the marathon distance this pacer uses?
It uses 42.195 kilometres exactly, the distance fixed by World Athletics. The imperial statement of 26 miles 385 yards works out to 42.194988 kilometres, twelve millimetres short of the metric figure, while the popular shorthand 26.2 miles is 42.1648128 kilometres, about 30 m short and worth roughly nine seconds at three-hour-thirty marathon pace. Every split here is computed from the metric definition and converted with the exact international mile of 1.609344 kilometres.
Sources behind the pacing and equivalence maths
Sources: Riegel, P. S. Athletic records and human endurance, American Scientist 69(3):285–290, 1981 (the T2 = T1(D2/D1)1.06 endurance relation); indexed at PubMed PMID 7235349. World Athletics, Marathon discipline page, which states the distance as 26 miles 385 yards (42.195 km): worldathletics.org. World Athletics in co-operation with AIMS, The Measurement of Road Race Courses, for the calibrated bicycle method, the 1.001 short course prevention factor and the 0.1 % (42 m) measurement uncertainty limit: aims-worldrunning.org. Daniels, J. and Gilbert, J., Oxygen Power: Performance Tables for Distance Runners (1979) and Daniels, J., Daniels' Running Formula (Human Kinetics), for the E, M, T, I and R training-zone vocabulary. Buist, I. et al., No effect of a graded training program on the number of running-related injuries in novice runners: a randomized controlled trial, American Journal of Sports Medicine 36(1):33–39, 2008: PubMed PMID 17940147. General activity guidance from the American College of Sports Medicine and the US Physical Activity Guidelines for Americans. Note on sourcing: the full text of the 1981 Riegel article was not retrievable online during preparation of this page, so the 1.06 exponent and its commonly quoted validity window of roughly 3.5 to 230 minutes are reported as published and widely reproduced rather than transcribed from the original.
Stride Symphony Challenge
Rehearse the split sheet you just generated. The ghost runner holds the quarter-by-quarter target pace from your plan; keep the orange runner alongside it as headwinds, downhills and tempo lifts try to pull you off rhythm.